Can screening for volatility prevent plunging portfolios?

Friday, Nov 02 2018 by
Can screening for volatility prevent plunging portfolios

Smarter searching for all-weather stocks

At the start of October, the FTSE All-Share fell by nearly 7% in a week. Good companies are still cheaper than they were.

But how best to find them?

October 2018 was a story of fear overcoming greed, with investors haunted by the spectre of a bear market. Very appropriate for Halloween. We have built up a loose body of commentary over the past couple of weeks, when not hiding under the bedsheets. Once the dust settles we can forget how things felt in the heat of the moment, so it’s helpful to have a record showing that, ah, yes, at one point, some of us - maybe - were a tad concerned.

Acknowledging these things, while not flattering to the ego, can better prepare us for the future. It will happen again, after all. Here are some of the articles and comments (from most recent to oldest):

Tuesday 30th October SCVR

Monday 29th October SCVR

Thursday 25th October: How falling markets present tough questions to investors

Thursday 25th October SCVR

Friday 19th October: Stable stock ideas for volatile markets

Anyway, back to identifying those investment opportunities. We need to narrow the field. One method of investing comes to mind.

Blitz, van Vliet, and the low volatility anomaly

  • There is a growing body of academic work that supports the ‘low volatility anomaly

  • It shows that low-volatility stocks tend to have higher returns per unit of risk than higher-volatility stocks

  • This begs the question: are we better off being boring?

There is a method of investing out there that has been used by some of the best fund managers in the world. Its successful application used to require many man-hours and entire teams of investment professionals due to its quantitative nature and the calculations involved. We have set up our own easy-to-use take on it that and, so far, as we see below, it has yielded encouraging results.

I’m talking about the low volatility anomaly. There is a growing body of academic studies that proves it works - here are a few of the more influential articles:

Unlock this article instantly by logging into your account

Don’t have an account? Register for free and we’ll get out your way


As per our Terms of Use, Stockopedia is a financial news & data site, discussion forum and content aggregator. Our site should be used for educational & informational purposes only. We do not provide investment advice, recommendations or views as to whether an investment or strategy is suited to the investment needs of a specific individual. You should make your own decisions and seek independent professional advice before doing so. The author may own shares in any companies discussed, all opinions are his/her own & are general/impersonal. Remember: Shares can go down as well as up. Past performance is not a guide to future performance & investors may not get back the amount invested.

Do you like this Post?
48 thumbs up
0 thumbs down
Share this post with friends

33 Comments on this Article show/hide all

Richard Goodwin 6th Nov '18 14 of 33

In reply to post #414829

I admire and respect your posts Nick and think they add real value to the bb. The main conclusion is always clear but I really struggle to understand some of the maths!

| Link | Share
Richard Goodwin 6th Nov '18 15 of 33

In reply to post #415084

But if one can successfully achieve the level of market timing necessary to make your suggested strategy work @dfs12 then why not just move between high vol and cash?

| Link | Share | 1 reply
skinner66 6th Nov '18 16 of 33

all these graphs are good but it didnt predict 2008 crash.. ive learnt to not panick now , after much panicking lol

| Link | Share
dfs12 6th Nov '18 17 of 33

In reply to post #415844

That sounds like a good plan. I reckon the performance would be pretty good. We just need a reliable market timing device to fortell the future!

| Link | Share
allonso 7th Nov '18 18 of 33

Beyond my understanding

| Link | Share
Trident 7th Nov '18 19 of 33

Looking at my own portfolio, I am sure some of the Risk categories of my selected stocks have drifted over time. Now they all labelled adventurous. This is a slippery business, as there is currently no ascribed timeline to any risk status. For example, if you were to screen for this characteristic, wouldn't you want to know how long it had fallen into that category?

Maybe its better to add to this other factors - average yield etc, or maybe screening as screening on one characteristic alone is no to be recommended?

I am a naive investor, and certainly not quant minded, as you may have gathered.

| Link | Share | 1 reply
goodgoff 7th Nov '18 20 of 33

Up to now I have been weeding out stocks that don't fall in the green sector of the bubble charts in Overview.
The 58 qualifying conservative stocks that currently meet your screen rules seem to fill the whole spectrum.
If I add QV, VM and QM filters at >80% then only 5 stocks appear.
I now ask myself which approach is more likely to protect my wealth?

| Link | Share | 1 reply
Jack Brumby 8th Nov '18 21 of 33

In reply to post #416504

That makes sense Trident - I don't know if you create screens already, but it's easy (and actually quite fun) to play around with them on Stocko as you can let out your inner mad scientist. I focused on RiskRatings here as I wanted to deepen my understanding of the concept, and including other factors in the screen might have dragged me off topic. But for sure, including some other measures to see if you can better tailor a screen to find the stocks you're looking for sounds like a good idea.

The StockRankitself is a composite measure of various ratios so that number contains quite a lot of info about a stock. Have you checked out the O'Shaughnessy book 'What Works on Wall Street'?

| Link | Share
Jack Brumby 8th Nov '18 22 of 33

In reply to post #416784

Hi goodgoff,

I don't think there's a wrong or right answer to this - not one I could suggest without a lot more data to back it up, anyway.

For protecting wealth, considering good, cheap stocks (the green area of the quality vs. value bubble chart) sounds like a good strategy. As does paying attention to the RiskRatings, which measures how volatile the shares have been on average over a multi-year period.

You could relax the screen slightly. I'm assuming you're looking just in the UK? If you drop the 'VM' and add in 'Balanced' that widens the universe of stocks a little.

If you're not comfortable doing that then don't, though. It depends what your own level of risk tolerance is. I just tried to find a Buffett quote (couldn't) but he's saying something like 'imagine you could only buy 10 stocks in your life' - the point being if I want to be very disciplined and very safe, I might have to bide my time and sit on cash waiting for the right opportunity to come along.

And even with the RiskRatings and StockRanks, it always pays to DYOR - does this company have a strong economic moat? Does it generate strong returns on capital and/or equity? If it pays dividends, does it have a history of sustainable payments with good dividend cover or has it had to stop divis or reduce them before? Has it survived through tough markets before?

I also remind myself that in October a) most stocks were down as, probably, were most investors and, b) share prices can recover (even if the present feels painful) - February 2018 was also quite a steep drop but it seemed like that pain was forgotten as soon stocks recovered heading into the summer.

That turned into a bit of a ramble!

| Link | Share | 1 reply
dangersimpson 8th Nov '18 23 of 33

Hi Jack,

Good to have you writing for Stockopedia.

As you say the low volatility 'anomaly' has been known since the days of Haugen and is a significant challenge to the EMH. Haugen proposed that it could be explained as a function of gearing limits on mutual funds combined with the principal-agent problem related to fund manager bonuses. i.e. FM get a bonus above a certain benchmark but doesn't pay it back in down years. Rational FM's with this incentive structure should prefer more volatile stocks even if they bid them up to prices which makes them overvalued, on average. They are in effect using them as an expensive form of gearing in order to maximise their personal expected return.

What truly matters in your analysis is out-of-sample performance. While it's not entirely clear from your article it reads as if you are taking today's RiskRating and looking at the historic 6 month volatility & return. Since the last 6 months have sen a negative FTSE return and the RiskRatings are formed form historic 12 month volatility it is no surprise that you get the results that you do since the formation data is in the sample whose results you are reporting.

What you really need to look at is the 6-month performance of the groups of Riskratings formed 6 months ago. And then as others have pointed out repeat for multiple periods.

I'm not sure you will see a low-volatility strategy outperform full cycle on an absolute basis but I'm pretty sure you will on a volatility adjusted basis, which is still a big positive since you can just gear up to generate absolute out-performance if you want, or without gearing it makes it a lot easier to hold equities full cycle if you reduce the volatility and is worth the sacrifice in absolute performance for most people (who tend to have negative-impact market timing tendencies!)


| Link | Share | 1 reply
Gromley 8th Nov '18 24 of 33

In reply to post #417054

Hi Jack,

I think the quote you were looking for is :

An investor should act as though he had a lifetime decision card with just twenty punches on it.

To me that goes hand in hand with the better known :

Diversification is a protection against ignorance. It makes very little sense for those who know what they’re doing.

The key message essentially being "don't invest unless you are absolutely sure."

In my view though that is very poor investment advice for the majority of investors and in fact it doesn't align very well with other things Buffett has said, such as :

We’ve made lots of mistakes, but they don’t bother me. We’ve had no regrets. We are in the business of making many decisions and there are bound to be mistakes.
Everyone is different of course, but for me trying to make a very small number of investments all of which must be correct is just way too much pressure in many ways and can in fact lead to worse not better decision making.

One of the eye openers for me (or at least reinforcement of something I probably knew anyway) from using Stockopedia is the realisation that if I can distil down my investment universe to a high quality galaxy, then I'll find enough planets rich in dilithium crystalsto make up for the dusters.

I certainly don't set out to make dumb investment decisions (and in fact do focus a lot on avoiding them), but I very much like not having to make genius investment decisions that always have to be right.

Live long and prosper!

| Link | Share | 1 reply
Dickyfit 9th Nov '18 25 of 33

An interesting article although, as others have pointed out, any analysis of data should use prospective rather than historic data. It is actually very easy to see how the SR and its different composites and components have performed recently as the markets have gone backwards using the Ranks Performance tool by just limiting it to the past 3 months. Value seemed to perform best looking at 50m+ companies UK?EU and I was surprised how well momentum held up. But there was no magic bullet. It would be interesting if Stocko would enhance the Rank performance tool to also allow slicing and dicing by Risk Rating and Style as well to facilitate the types of analysis that you have been doing.

| Link | Share
dsumners 13th Nov '18 26 of 33

Interesting article Jack. I'd be interested in seeing the Risk Ratings performance on a more regular basis. Would it be possible to do this on say a quarterly basis?

| Link | Share | 1 reply
pka 13th Nov '18 27 of 33

A very easy way to construct a well-diversified low volatility portfolio would be to buy an ETF such as this one, which invests globally in stocks exhibiting the "low volatility" factor:

It has performed reasonably well over the past 5 years, albeit during a mainly bull market, but not as well over the past 3 years as its corresponding global "momentum factor" ETF:

The following web articles suggest that buying both a low-volatility factor ETF and a momentum factor ETF, with occasional rebalancing between them, would provide good long-term performance with lowish volatility, because low-volatility stocks often move in the opposite direction to momentum stocks during bear market periods:

| Link | Share | 1 reply
Nick Ray 13th Nov '18 28 of 33

In reply to post #418539

Funnily enough I was just thinking about this myself. The momentum strategy is high beta and so it does well when the market is bullish and the low volatility strategy is low beta and is better in a downturn.

But the argument against is that the market volatility should already be the the lowest volatility per unit of return solution already, and in the graphs it does look as if the combination more-or-less cancels out.

| Link | Share
Jack Brumby 14th Nov '18 29 of 33

In reply to post #417119

Haha thanks Gromley, I feel like I've just watched a sci-fi film. Made me chuckle! A toast to planets rich in dilithium crystals.

I generally agree about the Buffett quote (thanks for locating it btw) in that it is an unrealistic extreme. I wouldn't take it literally and hopefully others don't. It does pop into my head from time-to-time as more of a fable when I'm on the fence about making an investment though.

I'm trying to be mindful of people with different time horizons and risk tolerances, as well. Eg. if you are going to invest for 3-5 years max and your priority is just to not lose money, then simply parking your cash in the bank might not be such a bad thing. Obviously this won't apply for the majority of investors as you say.

See you around!

| Link | Share
Jack Brumby 14th Nov '18 30 of 33

In reply to post #417094

Hi Mark,


I take your point about out-of-sample data and the potential for hindsight bias. Others have raised it as well and I'm definitely taking it on board for studies in future. Volatility is a big theme and it's not going to go away, so we are definitely going to keep coming back to it. This should give us the opportunity to improve and refine the way we can look at it.

'negative-impact market timing tendencies' - that's a nice way of describing those moments of folly we all know and love!

| Link | Share | 1 reply
Jack Brumby 14th Nov '18 31 of 33

In reply to post #418524

This article was a lot of fun so I'd like to! At the very least, I think we will try to release more content on the RiskRatings as volatility is such a big theme to cover. Regular periodic coverage could be a good way to do it...

| Link | Share
Andrew L 14th Nov '18 32 of 33

Personally I think using volatility as a measure of risk is not the best approach. It does have some use as miners are volatile and are high risk.

But Amazon is surely lower risk than say a bricks and mortar retailer. But with Amazon a highly rated growth company its shares have been volatile. The trouble with using volatility is that it penalizes companies that have more of their earnings value in the future. But these can be some of the best long-term investments.

You are also relying on other investors to identify risky stocks for you i.e. through the volatility. Then you are assuming that risk can be forecast. The whole point of risk most of the time is that we don't see it coming.

I don't think it is rocket science to say which companies are risky on the basis of what they do and some financial metrics. Some companies with low volatility can blow up in short measure.

| Link | Share
dangersimpson 14th Nov '18 33 of 33

In reply to post #418794

Hi Jack,

Yes volatility is one of the most important topics in investing, and there are some big questions such as, to what extent does it represent risk to an investor? How persistent is volatility? How does it relate to other risk or behavioural factors? Stockopedia should increasing start to have the data sets available to start to answer these questions so I look forward to your future research in this area.

Generally in-sample data for a result that isn't used in your formation data set may be ok, although any positive results could be due to data mining and following up with out of sample data is usually best to try to rule that out. The big issue here is that your result (lower 6-month volatility) is actually part of your formation data set (1-year volatility.) It is much more important to think through the independence of formation & result sets, particularly since many investing metrics rely on the same accounting or price data, rather than just shift to out-of sample data.


| Link | Share

Please subscribe to submit a comment

Stock Picking Tutorial Centre

Let’s get you setup so you get the most out of our service
Done, Let's add some stocks
Brilliant - You've created a folio! Now let's add some stocks to it.

  • Apple (AAPL)

  • Shell (RDSA)

  • Twitter (TWTR)

  • Volkswagon AG (VOK)

  • McDonalds (MCD)

  • Vodafone (VOD)

  • Barratt Homes (BDEV)

  • Microsoft (MSFT)

  • Tesco (TSCO)
Save and show me my analysis